29th Symposium on Fusion Technology (SOFT 2016)

O4C.3 Qualification of MELCOR and RELAP5 nodalization models for EU HCPB and HCLL TBS accident analyses

8 Sep 2016, 11:40

20m

Meeting Hall V 2nd floor (Prague Congress Centre)

Board: 3

Oral J. Power Plants Safety and Environment, Socio-Economics and Technology Transfer O4C

Speaker

Andrew Grief (Amec Foster Wheeler)

Description

'Fusion for Energy' (F4E) is designing, developing, and implementing the European Helium-Cooled Lithium-Lead (HCLL) and Helium-Cooled Pebble-Bed (HCPB) Test Blanket Systems (TBSs) for ITER. Safety demonstration is an essential element for the integration of these TBSs into ITER and accident analysis is one of its critical components. The F4E, Amec Foster Wheeler and INL comprehensive methodology for fusion breeding blanket accident analysis, published last year, consists of several phases. The methodology starts with the selection of reference accident scenarios, the development of detailed accident analysis specifications and the assessment of analysis codes. Models of each TBS are then constructed using the selected codes (MELCOR 1.8.5 and RELAP5-3D for the HCLL TBS) and modelling approaches. The models are qualified according to a test matrix including comparison with TBM finite element design analyses, code-to-code comparisons (between the MELCOR 1.8.5 and RELAP5-3D models) for both TBS normal operation and transient cases, and sensitivity studies for accident scenarios. The qualification test cases that are executed gradually move from models of separate systems to complete TBS models, and from the simulation of steady-state and normal plasma pulse operation to consideration of power excursions, operational transients and accident events. Finally, both of the qualified models are used to analyse a selected accident scenario (a 32 hour loss-of-offsite power) together with sensitivity studies dedicated to the evaluation of uncertainties. This step completes the qualification process. The impact of uncertainties associated with the accident analyses is also addressed to provide confidence in the level of conservatism in the results. Following an expert review of areas of uncertainty (including phenomena identification and ranking table (PIRT)) a gradual approach to uncertainty assessment has been adapted. The results obtained in the qualification of the EU HCLL and HCPB TBS models and their uncertainty evaluation will be reported in the paper.